Placental Alkaline Phosphatase Promotes Zika Virus Replication by Stabilizing Viral Proteins through BIP

ZIKV is a recently emerged mosquito-borne flavivirus that can cause devastating congenital Zika syndrome in pregnant women and Guillain-Barré syndrome in adults, but how ZIKV specifically targets the placenta is not well understood. Here, we identified an alkaline phosphatase (ALPP) that is expressed primarily in placental tissue and promotes ZIKV infection by colocalizing with ZIKV proteins and preventing their proteasome-mediated degradation. The phosphatase activity of ALPP could be required for optimal ZIKV infection, and ALPP is stabilized by BIP via its chaperone activity. This report provides novel insights into host factors required for ZIKV infection, which potentially has implications for ZIKV infection of the placenta.

membrane. We performed a viral entry assay using JEG-3 and HK-2 cells as previously described (28). Although ALPP depletion did not affect the entry of ZIKV ( Fig. 2B; see also Fig. S4A), ZIKV vRNA synthesis was significantly reduced in ALPP Ϫ/Ϫ JEG-3 cells at 12 and 24 h postinfection (Fig. 2C), which suggested that ALPP likely functions at postentry steps. Furthermore, we analyzed the production of ZIKV SZ01 and MR766 virions from infected wild-type (WT) and ALPP Ϫ/Ϫ JEG-3 cells over an infection period of 72 h and found that ALPP depletion markedly reduced the release of ZIKV virions ( Fig. 2D; see also Fig. S4B). Consistent with the results from JEG-3 cells, the numbers of ZIKV plaque formation units were also significantly reduced in the supernatant of ALPP-KO U-251 MG cells (Fig. S4C). Immunoblot analyses showed that expression of ZIKV proteins, including NS1, NS2B, and NS3, was nearly absent in ALPP Ϫ/Ϫ JEG-3 cells 24 h postinfection relative to that in WT JEG-3 cells (Fig. 2E). These results further suggest that ALPP is essential for vRNA synthesis and productive ZIKV infection in human placental trophoblasts and astrocytes.
ALPP stabilizes viral proteins during ZIKV infection. Following ZIKV entry, viral (ϩ)RNAs are released and then translated to produce viral proteins (30,31). The nonstructural proteins form replication complexes on modified ER membranes and transcribe genomic (ϩ)RNA into a negative-strand RNA, which serves as the template for the production of progeny (ϩ)RNAs. Because the replication of (ϩ)RNA viruses involves numerous interactions between components from the virus (RNA and proteins) and the host (proteins, membranes, and lipids), we hypothesized that ALPP might promote ZIKV vRNA synthesis by participating in the initial production of ZIKV proteins. We examined whether the expression levels of nonstructural proteins, including components of the viral replicase (NS2B, NS4A, NS4B, and NS5) and the assembly complex (NS1) (32), were dependent on ALPP. We transiently overexpressed the nonstructural proteins of ZIKV in WT and ALPP Ϫ/Ϫ JEG-3 cells. Interestingly, we found significant reductions in the levels of ZIKV NS1, NS2B, NS4B, and NS5 in ALPP Ϫ/Ϫ JEG-3 cells Capital letters in parentheses correspond to panels B to E of the figure. (B) Detection of ZIKV vRNA by RT-qPCR. Statistical analyses of internalized ZIKV were performed; MOI ϭ 100. The data were obtained from four biological replicates (n ϭ 4). n.s., no significance (P ϭ 0.1934; two-tailed Student's t test). (C) Detection of intracellular ZIKV vRNA at the indicated times after ZIKV infection by RT-qPCR. MOI ϭ 0.5. The data were obtained from four biological replicates (n ϭ 4). The significant differences between WT and ALPP Ϫ/Ϫ JEG-3 cells at each time point were determined by two-way ANOVA with Sidak's multiple-comparison test. ****, P Ͻ 0.0001; n.s., no significance; ND, not detected. (D) ZIKV MR766 (left) and SZ01 (right) infectious particles produced from WT and ALPP Ϫ/Ϫ JEG-3 cells at the indicated time after ZIKV infection. MOI ϭ 0.5. The data were obtained from at least three biological replicates (n Ն 3). Significant differences between WT and ALPP Ϫ/Ϫ cells at each time point were determined by two-way ANOVA with Sidak's multiple-comparison test. *, P Ͻ 0.05; **, P Ͻ 0.01; ***, P Ͻ 0.001. TCID 50 , 50% tissue culture infective doses. (E) Production of ZIKV MR766 (top) and SZ01 (bottom) proteins in WT and ALPP Ϫ/Ϫ JEG-3 cells infected with ZIKV for 24 h. MOI ϭ 0.5. The quantitative data in this figure are shown as means Ϯ SEMs.
( Fig. 3A) but minimal reduction in the NS4A expression level, which suggests that ALPP directly affects individual ZIKV proteins even in the absence of integral components of the viral machinery. In contrast, the protein levels of IAV NP and PB2 in ALPP Ϫ/Ϫ JEG-3 cells were not reduced compared to those in WT cells (Fig. S5A). Notably, treatment of ALPP Ϫ/Ϫ JEG-3 cells with MG132, a proteasome inhibitor, rescued the expression of NS1, NS2B, NS4B, and NS5 (Fig. 3B). These data indicate that in the absence of ALPP, the nonstructural proteins of ZIKV are targeted for degradation, which suggests that ALPP actively stabilizes these proteins during ZIKV infection.
To corroborate the hypothesis that ALPP could promote ZIKV replication by stabilizing viral replicase to prevent its degradation, we subsequently examined the observations mentioned above during ZIKV infection. We established a U-251 MG cell line stably expressing Flag-tagged ALPP, isolated a cell clone (clone 15) with a high level of ALPP expression by fluorescence-activated cell sorter (FACS) analysis (Fig. S5B), and then performed coimmunoprecipitation experiments in the context of ZIKV infection. Immunoblot analyses revealed that ALPP interacted with the ZIKV nonstructural proteins NS1, NS2B, NS3, and NS5 and with the structural proteins capsid (C) and envelope (ENV) (Fig. 3C). Moreover, ALPP bound to the NS5 and envelope (E) proteins after the corresponding genes were transiently introduced into the cells (Fig. S5D). Immunofluorescence analysis of ZIKV-infected Flag-ALPP-positive cells showed that cytoplasmic ALPP Promotes Zika Virus Replication ® ALPP colocalized with ZIKV NS2B, capsid, and ENV (Fig. 3D). Taken together, these results suggest that ALPP plays an important role in ZIKV infection through its direct interaction with ZIKV proteins.

BIP associates with ALPP and is essential for ZIKV infection.
To determine whether additional host factors participate in the ALPP-ZIKV interactome, we performed a coimmunoprecipitation experiment with uninfected Flag-ALPP-overexpressing U-251 MG cells and pulled down a distinctive 78-kDa protein (Fig. S6A). Mass spectrometry analyses identified this protein as BIP (also called GRP78), an ER-resident Hsp70 chaperone that assists in protein folding and in the surveillance of misfolded proteins (27) (see Table S1B in the supplemental material). ALPP is directed to the ER via an N-terminal cleavable signal peptide. At its extreme C-terminal end, a hydrophobic stretch of amino acids enables attachment of the GPI anchor and is subsequently cleaved off. A previous study (26) demonstrated that BIP binds to ALPP specifically via its C-terminal hydrophobic stretch. BIP thus binds only the proform of ALPP and not the mature GPI-linked form. To assess whether the GPI anchor or the hydrophobic stretch at the C terminus is required for the function of ALPP in ZIKV infection, we constructed an ALPP-C-terminus-deleted protein (ALPP-ΔC33) and an ALPP-D484W mutant that rendered a loss of the GPI linker (Fig. 4C). Deletion of 33 amino acids at the C terminus significantly inhibited the function of ALPP in ZIKV infection (Fig. 4D). In contrast, the GPI-defective mutant (D484W) did not affect the function of ALPP in ZIKV infection ( Fig. 4D), which indicated that the addition of a GPI anchor (mature form) is not required for ALPP function in ZIKV infection.
To confirm the interaction between BIP and ALPP, we immunoprecipitated ectopically expressed Flag-tagged ALPP and found that BIP was indeed pulled down both in the absence (Fig. 5A, left) and in the presence (Fig. 5A, right) of ZIKV infection. This finding was further corroborated by those in Flag-ALPP-293T cells ( Fig. S5C and S6A) and Flag-ALPP-JEG-3 cells ( Fig. S6A and B). Furthermore, the immunofluorescence of Flag-tagged ALPP in U-251 MG cells revealed a reticular pattern of ALPP localization that overlapped that of BIP (Fig. 5B). These results indicate the participation of BIP in the ALPP-viral protein interactome.
To evaluate the role of BIP, we deleted BIP in U-251 MG cells via CRISPR/Cas9 analysis (Fig. 5C) and observed a significant reduction in ALPP expression in BIP -/-U-251 MG cells (Fig. 5D). However, the ALPP mRNA levels were unaffected in BIP -/cells (Fig. 5E). To further dissect the influence of ALPP on BIP, we examined the expression of BIP in WT and ALPP -/-JEG-3 cells and found that ALPP depletion had no effect on BIP expression (Fig. 5F). However, despite their normal expression of BIP, ALPP-depleted cells showed significantly reduced ZIKV NS1 protein expression (Fig. 5F). Moreover, the knockdown of BIP decreased the endogenous ALPP level and reduced ZIKV NS1 protein expression (Fig. 5G).
Cochaperones, such as DNAj-like proteins, are known to cooperate with BIP to stabilize and fold proteins (20). Therefore, we wanted to determine if ALPP may be acting as a cochaperone. We aligned ALPP with other DnaJ-like cochaperone sequences and found that all DnaJ-like cochaperones, but not ALPP, contained a J-domain sequon (KYHPDK) ( Fig. S7A and B). Molecular docking results suggested that no similar structural domains exist in ALPP and DnaJ-like cochaperones (Fig. S7C). Based on these findings, ALPP might not function as a DnaJ-like cochaperone.
BIP chaperone activity is essential for ZIKV infection and ALPP stability. BIP has been implicated in the viral life cycle of yellow fever virus (YFV) and Japanese encephalitis virus (JEV) (34,35). Here, we demonstrated that BIP depletion significantly reduced ZIKV infection but did not affect DENV infection (Fig. 6A) (20). However, the entry of ZIKV into cells was not influenced by BIP depletion (Fig. S6C), which is similar to the results found in ALPP Ϫ/Ϫ cells. Furthermore, the expression of viral proteins, including ENV, PrM, and NS1, was significantly reduced in BIP -/-U-251 MG cells 24 h post-ZIKV infection (Fig. 6B) and 24 h posttransfection with ZIKV proteins (Fig. S6D). These results suggest a specific role of BIP in ZIKV infection and in the stability of ZIKV proteins. We subsequently examined whether BIP also interacts with viral proteins through coimmunoprecipitation experiments using ZIKV-infected Flag-BIP-positive 293T cells (Fig. S8A) and U-251 MG cells (Fig. S8B). Similarly to the observations obtained for ALPP, we found that BIP exhibited associations with the ZIKV nonstructural proteins NS1, NS3, NS2B, and NS5 and the structural proteins ENV and C.
To test whether BIP chaperone activity is required for its regulation of ALPP and ZIKV infection, we reconstituted BIP-depleted cells with CRISPR-resistant wild-type BIP or ATPase-defective mutants (36). The mutants could not rescue the ALPP levels ( Fig. 6C) or ZIKV infection (Fig. 6D), which suggested that BIP chaperone activity is required for ZIKV infection. Furthermore, the interaction of ALPP with ZIKV proteins or BIP was abrogated by the addition of ATP, which binds to the chaperone and induces a conformational change that releases substrates (Fig. 6E) (37).

DISCUSSION
Clinical observations suggest that ZIKV infection in pregnant women can restrict fetal growth and cause other fetal congenital abnormalities (1, 9, 10, 13). However, to access the developing nervous system and disrupt normal neuronal development, ZIKV must penetrate and replicate in the placenta. Indeed, in vitro studies have shown that ZIKV can infect maternal decidual cells, first-trimester cytotrophoblasts, and extravillous trophoblasts (11,38,39). Previous studies have also provided evidence showing positive ZIKV replication in the placenta, the brain tissues of fetuses with microcephaly, and placenta-derived primary cells (3,4,6,(8)(9)(10)(11). ALPP is primarily expressed in placental and endometrial tissues, and its expression is further upregulated during pregnancy (23)(24)(25). Here, we found that ALPP may promote high levels of ZIKV but not ALPP Promotes Zika Virus Replication ® DENV infection by stabilizing both nonstructural and structural viral proteins together with the chaperone BIP. High levels of ZIKV activity have been detected in human and mouse placentas (3-5, 7, 8), and these are likely promoted by high levels of placental ALPP. Additionally, ALPP might also enhance ZIKV replication in other tissues, as exemplified by our findings with the U-251 MG and HK2 cell lines, which were previously confirmed to be highly permissive to ZIKV infection (28,40). Together, our results provide evidence suggesting that ALPP exclusively facilitates ZIKV infection in placental trophoblasts and fetal brain cells, based on data obtained with JEG-3 and U-251 MG cells.
ALPP is a homodimeric enzyme, and each catalytic site contains three metal ions (two Zn ions and one Mg ion) that are necessary for enzymatic activity. Mutations of the enzymatic activity sites affect the hydrophobic pocket formation of ALPP and its function (33). Intriguingly, we found that the phosphatase activity of ALPP was likely required for optimal ZIKV activity. Previous studies uncovered that differentially phosphorylated states of flavivirus NS5 affected its subcellular localization and interaction with NS3, controlling its function in the viral RNA replicase (52,53). Moreover, dephosphorylation of West Nile virus capsid protein enhances the processes of nucleocapsid assembly (54). Thus, we hypothesize that the phosphatase activity of ALPP may control the phosphorylation status of ZIKV proteins and may regulate the aggregation, interaction, stability, and function of ZIKV proteins in viral RNA replicase. Dynamic phosphorylation of ZIKV structural proteins may also modulate viral assembly and maturation. Finally, the enzymatic activity of ALPP could also facilitate its proper structural conformation and interaction with BIP, which stabilizes ALPP during ZIKV infection. In the future, more studies are needed to further understand how ALPP and its enzymatic activity precisely regulate ZIKV life cycle.
The proper folding and assembly of viral proteins might be mediated by host chaperones. Previous studies have shown that cytosolic Hsp70 isoforms (including HSPA1A, HSPA1B, and HSPA8) are needed at distinct steps of the flavivirus viral cycle, including entry, RNA replication, and biogenesis of virions (20,43). During JEV infection (55), BIP is needed for entry and formation of the replication complex. In this study, we demonstrated that BIP cooperates with ALPP and plays roles exclusively in ZIKV infection and not in DENV infection. Given the high degree of similarity between flavivirus members, the requirement of distinct isoforms of Hsp70 in various viral functions is intriguing. In this study, we identified a specific BIP-ALPP partnership due to the strong and specific binding of these proteins in placental cells. Our data further demonstrate that the specific requirement of BIP by ZIKV is likely endowed by its partner, ALPP. In fact, in the absence of ALPP, BIP itself cannot ensure the stability of ZIKV proteins and fails to support ZIKV infection. Thus, our findings provide a new model regarding how chaperone proteins can function specifically during flavivirus infections. Interestingly, antibodies directed against both the N and C termini of BIP suppress both the binding of the DENV to the cell surface and the infectivity of this virus in HepG2 cells (56). Our data raise the possibility that ALPP or BIP can be targeted to intervene in ZIKV infection.
Taken together, the results indicate that identification of the host factors that play important roles in regulating the ZIKV life cycle provides novel therapeutic targets for the development of antiflavivirus drugs. ALPP might be targeted to block ZIKV infection in the placenta and to ameliorate microcephaly. These possibilities should be explored further in a mouse model. An in-depth understanding of virus mechanisms and biology will facilitate the development of antiviral antibodies, vaccine therapies, and virustargeted small molecules.

MATERIALS AND METHODS
Cell lines and viruses. Aedes albopictus C6/36 cells were grown in a mixture of 30% RPMI 1640 (Gibco) and 60% Dulbecco's modified Eagle's medium (DMEM; Gibco) supplemented with 10% fetal bovine serum (FBS; Gibco). The U-251 MG cell line was purchased from BeNa Culture Collection (BNCC), and the 293T, HK2, and JEG-3 cell lines were purchased from American Type Culture Collection (ATCC). The 293T and U-251 MG cells were cultured in DMEM (Gibco) supplemented with 10% FBS, 100 IU/ml penicillin, and 100 g/ml streptomycin. The BIP-KO cells were cultured in DMEM (Gibco) supplemented with 15% FBS, and the JEG-3 cells were cultured in minimal essential medium (MEM) (Gibco) supplemented with 10% FBS, 100 IU/ml penicillin, and 100 g/ml streptomycin. The HK2 cells were cultured in DMEM/F12 (Gibco) (1:1) supplemented with 10% FBS, 100 IU/ml penicillin, and 100 g/ml streptomycin. The 293T, U-251 MG, JEG-3, and HK2 cells were maintained at 37°C, and the C6/36 cells were maintained at 28°C in a fully humidified atmosphere containing 5% CO 2 . All the cell lines were tested by Saily Bio (Shanghai, China) and were free of mycoplasma contamination. The ZIKV MR766 stock was purchased from ATCC (ATCC VR-1838). The ZIKV SZ01 virus strain (GenBank accession number KU866423) was kindly provided by Cheng-Feng Qin at the Department of Virology, State Key Laboratory of Pathogen and Biosecurity, Beijing Institute of Microbiology and Epidemiology.
CRISPR-Cas9 cloning. Oligonucleotides encoding sgRNAs for generating KO cells using CRISPR-Cas9 were cloned into lentiCRISPRv2 plasmid (Addgene Plasmid, 52961) as previously described (28). The oligonucleotide sequences of the sgRNAs targeting ALPP and BIP are listed in Table S1B in the supplemental material. LentiCRISPRv2 clones containing the guide sequences were sequenced and purified and used for lentiviral production. To generate heterogeneous KO cell populations, JEG-3, U-251 MG, or HK2 cells were infected with the lentiCRISPRv2-derived lentivirus for 36 h and then reseeded into complete DMEM containing 1 to 4 g/ml puromycin for 14 days to select for the transduced cells. Single-cell clones of JEG-3 cells targeted for CRISPR-mediated ALPP KO were diluted with the parental JEG-3 cells at a ratio of 1/1,000. One of the cells from the cell mixtures was plated in each well of a 96-well plate. Once the cells reached confluence, the cells were passaged based on a 48-well format in the presence of 4 g/ml puromycin to kill nontargeted cells. The surviving populations derived using this approach were propagated and expanded for 4 weeks prior to cryopreservation of the stock cultures. Using this strategy, an ALPP-targeted KO cell line (ALPP Ϫ/Ϫ ) and a BIP-targeted KO cell line (BIP Ϫ/Ϫ ) were generated, and the gene KO efficiency was analyzed through immunoblot or immunofluorescence assays.
ZIKV infection. Prior to ZIKV infection, U-251 MG, JEG-3, and HK2 cells were seeded in 24-well plates (2 ϫ 10 4 cells per well) or 6-well plates (1 ϫ 10 5 cells per well). Twenty-four hours after seeding, the cells were rinsed once with phosphate-buffered saline (PBS) and then incubated with ZIKV at the indicated multiplicity of infection (MOI) in serum-free medium for 1 h at 37°C unless otherwise noted. The ZIKV-containing medium was then replaced with fresh DMEM or MEM supplemented with 2% FBS.
For the immunofluorescence assay, cells were seeded in four-or eight-well glass Millicell EZ Slide chambers (Millipore) at 4 ϫ 10 4 or 2 ϫ 10 4 cells per well and then infected with ZIKV as described above. After incubation at 37°C for the indicated times, the cells were rinsed twice with phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde (PFA).
For the plaque assay, 1 ϫ 10 5 Vero cells were seeded in 12-well plates. The next day, monolayer cells were infected for 2 h with serial dilutions of the supernatant collected from WT and ALPP-KO U-251 cells at 24, 48, and 72 h. The inoculum was then removed, and the cells were incubated in medium containing 2% FBS and 0.5% agarose. The cell monolayer was fixed with 10% formalin and stained with 1% crystal violet, and the plaques were counted after 5 days.
Immunoprecipitation assay. We used two approaches for the coimmunoprecipitation experiments. In the first approach, WT and ALPP-or BIP-expressing cells were mock infected or infected with ZIKV for 24 h. In the second approach, U-251-2ϫFlag control cells or U-251-ALPP-2ϫFlag cells (2 ϫ 10 6 cells per 10-cm-diameter dish) were transiently transfected with 14 g of pCMV-NS5 or pCMV-E separately using TurboFect transfection reagent (Thermo Fisher Scientific). The cells were rinsed twice with cold PBS, transferred to clean tubes, and lysed in cell buffer supplemented with 1% protease inhibitor cocktail (Sigma, catalog number P8340) for Western blotting and immunoprecipitation. The cell lysates were incubated with Pierce protein A/G agarose (Sigma, 20422) for 4 h at 4°C and then centrifuged at 10,000 ϫ g and 4°C for 10 min. When used, ATP was included at a final concentration of 2 mM (Selleck, S1985). The supernatant was transferred to a new tube and incubated with 30 l of anti-Flag M2 affinity gel (Sigma, A2220) overnight at 4°C. The Sepharose samples were centrifuged and then washed five times with cell lysis buffer and eluted using 3ϫFlag peptide (Sigma, F4799). All the samples were then boiled with SDS loading buffer for 10 min.
For ALPP and BIP detection, ALPP-U-251 MG cells were incubated first with mouse anti-ALPP antibody (R&D, MAB59051) (1:100) or rabbit anti-BIP antibody (ABclonal, A0241) (1:200) as appropriate and then with an Alexa Fluor-conjugated secondary antibody (AF-488 for visualization in the green channel and AF-568 for visualization in the red channel). All the cells were mounted with ProLong Gold Antifade with DAPI (4=,6-diamidino-2-phenylindole) (Life Technologies, P36931) and imaged using a TissueFAXS 200 flow-type tissue cytometer (TissueGnostics GmbH, Vienna, Austria). Some images were acquired with a Zeiss Axio Imager Z2 microscope (Wetzlar, Germany) using a 63ϫoil immersion lens objective and a Hamamatsu ORCA-Flash4.0 scientific complementary metal oxide semiconductor (sCMOS) camera (Hamamatsu City, Japan). All the statistical analyses of the immunofluorescence staining data were performed using the results from at least 3,000 cells per replicate, and the data are shown as means Ϯ standard errors of the means (SEMs). Viral entry assay. ZIKV cell entry experiments were performed based on a previously described protocol (28). In brief, for the virus binding assay, 1 ϫ 10 5 cells were seeded in a 12-well plate, cultured for 24 h, and infected with ZIKV (MOI ϭ 20) in cold MEM on ice for 1 h. Any unbound virus was removed, and the cells were then washed twice with 1ϫ PBS. The cells were incubated with prewarmed medium for 1 h at 37°C to initiate ZIKV internalization, and cell lysates were then harvested for vRNA quantitation by quantitative real-time PCR (RT-qPCR).
RT-qPCR. RNA was extracted using an RNeasy RNA isolation kit (Qiagen) according to the manufacturer's instructions. One microgram of RNA was transcribed into cDNA using random primers and Moloney murine leukemia virus reverse transcriptase (Promega, Charbonnieres, France). RT-qPCR was performed using the resulting cDNA templates and GoTaq qPCR master mix (Promega) with an Applied Biosystems 7300 real-time PCR cycler (see the supplemental material for the primer sequences). The PCR data were analyzed using SDS software (Applied Biosystems), and GAPDH (glyceraldehyde-3-phosphate dehydrogenase) expression was used as an internal control. For the ZIKV infection experiments, the expression of the genes of interest in the infected cells was normalized to that in unstimulated WT cells unless otherwise noted. All presented RT-qPCR data represent results from four biological replicates.
Statistical analyses. For quantification of the infected cells, at least 1,000 fluorescent cells were imaged and counted using a flow-type tissue cytometer. Three replicates were established per sample. The data are presented as means Ϯ SEMs. Significant differences between groups were determined using a two-tailed Student's t test unless otherwise noted. For the RT-qPCR analyses, at least four biological replicates were established for each sample. RT-qPCR-derived values that were 10-fold higher or lower than the mean values were identified as outliers and excluded from the analyses. The data are presented as means Ϯ SEMs, and significant differences between groups were determined by two-way analysis of variance (ANOVA) with the Sidak multiple-comparison test unless otherwise noted.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only.

ACKNOWLEDGMENTS
We thank Jia Liu (ShanghaiTech University) for providing the screening library plasmids.
This work was supported by the National Science Foundation of China (81901598 and 81430030) and the Intramural Funding from Shanghai Public Health Clinical Center (to J.C.).
We contributed to this work as follows. J.X. and J.C. conceived this study. J.C., S.Z., X.Z., and J.X. designed the experiments. J.C. and Z.C. performed the experiments and analyzed the data. D.F. and M.L. prepared the ZIKV. T.Q. and C.Z. contributed to the analyses and interpretation of the data. J.C., Z.C., S.Z., and J.X. drafted the manuscript. All of us contributed to the editing of the manuscript.
We declare that we have no competing financial interests.